1. Field of the Invention
The present invention relates to an image sensor pixel array, and more particularly, to a pixel array having a wide dynamic range and good color reproduction and resolution and an image sensor using the pixel array.
2. Description of the Related Art
An ideal image sensor reacts to light having an intensity of 0 lux. However, a real image sensor starts reacting to light having an intensity of a predetermined lux level more than 0. A starting point where the real image sensor starts reacting is called a minimum light intensity of a pixel. A point where the light intensity increases and the image sensor does not react any more is called a maximum light intensity of a pixel.
A dynamic range is defined as a range of reactions for a relative light intensity that a system can represent. Generally, the lowest limit of the dynamic range is limited by the minimum light intensity, and the supremum is limited by the maximum light. The image sensor cannot sense and represent the light that has more than the maximum light intensity.
FIG. 1 illustrates an output voltage of a pixel corresponding to a light intensity when an exposure time of a light detector is long and short.
Referring to FIG. 1, when a light detector included in an image sensor picks up an image signal with a short exposure time, a full line A represents responses of a pixel generating a pixel output voltage Data corresponding to the image signal (light). On the other hand, when the light detector picks up an image signal with a long exposure time, a dashed dotted line B represents a pixel output voltage corresponding to the image signal.
The full line A that is a response curve when the light detector picks up the image signal with a short exposure time is represented as a linear line having a gentle slope that increases until a value Data of the pixel output voltage corresponding to the image signal is saturated (referred to as saturation point) as an intensity of the image signal increases. On the other hand, the dashed dotted line B that is a response curve when the light detector has a long exposure time is represented as a linear line having a sharp slope that increases until the value Data of the pixel output voltage corresponding to the image signal is saturated as the intensity of the image signal increases.
Referring to the full line A, when the exposure time is short, image signals corresponding to regions {circle around (1)}, {circle around (2)}, {circle around (3)}, and {circle around (4)} can be converted into electric signals. Particularly, the image signals corresponding to the entire of the region {circle around (4)} where an intensity of the image signal is high can be converted into the electrical signals. Referring to the dashed dotted line B, when the exposure time is long, image signals corresponding to the regions {circle around (1)}, {circle around (2)}, {circle around (3)}, and {circle around (4)} can also be converted into electric signals. However, image signals having intensities corresponding to a region {circle around (5)} that is a part of the region {circle around (4)} where the intensity of the image signal is high have the same electric signals. In other words, the pixel has a disadvantage of being incapable of distinguishing the image signals having the intensities corresponding to the region {circle around (5)} from each other. However, the pixel has an advantage of more precisely representing changes in the intensity of the image signal corresponding to the region {circle around (1)} where the intensity of the image signal is low.
Accordingly, when a bright image signal is converted into an electric signal, picking up the image signal with the short exposure time (referred to as A) is more preferable. On the other hand, when a dark image signal is converted into an electric signal, picking up the image signal with the long exposure time (referred to as B) is more preferable.
Conventionally, in order to increase the dynamic range, the light intensities are classified into several regions according to exposure times and photographing is performed as follows. First, image frame data that is picked up with a short exposure time is stored in a memory.
Second, image frame data that is picked up with a long exposure time is stored in a memory.
Third, the two types of the frame data which are separated from each other by the several regions as illustrated in FIG. 1 and stored in the memory are properly combined to generate new image frame data having a wide dynamic range. Here, the first photographing and the second photographing are formed with the same image but different exposure times.
When the light intensities belong to the regions {circle around (2)} and {circle around (3)}, the pixel output voltage Data can be obtained by adding an amount B taken with a long exposure time and an amount A taken with a short exposure time and can be represented by Equation 1.D({circle around (2)})=xA+yB D({circle around (3)})=yA+xB  [Equation 1]
Here, D({circle around (2)}) represents a pixel output voltage when the light intensity belongs to the region {circle around (2)}, and D({circle around (3)}) represents a pixel output voltage when the light intensity belongs to the region {circle around (3)}. The sum of the variable x (x>0) and the variable y (y>0) is 1, and it is assumed that the variable x is smaller than the variable y (x<y).
Referring to Equation 1, when the light intensity belongs to the region {circle around (2)}, that is, it is dark (referred to as D({circle around (2)}), the applied amount B taken with the long exposure time is larger than the applied amount A taken with the short exposure time. On the contrary, when the light intensity belongs to the region {circle around (3)} that is brighter than the region {circle around (2)} (referred to as D({circle around (3)})), the applied amount A taken with the short exposure time is larger than the applied amount B taken with the long exposure time.
Conventionally, the image signals are converted into electric signals by using the two pieces of photographing information having different exposure times in order to increase the dynamic range of the image sensor. However, this method has disadvantages in that photographing has to be performed two times with different exposure times, it is not easy to classify the light intensity into the regions {circle around (2)} and {circle around (3)}, and it is not easy to determine a ratio (x, y) applied to the regions {circle around (2)} and {circle around (3)}. Due to the disadvantages, application of moving images is limited.
A color filter of a general image sensor uses a Bayer pattern using RGB (red, green, and blue) colors. In this case, a range of the RGB color filter for representing colors is limited than a range sensed by a person.
FIG. 2 illustrates color coordinates using a uniform color space defined by the CIE (International Commission on Illumination) in 1976.
Referring to FIG. 2, the region a indicates a range of colors sensed by an eye of a person, and the region b indicates a range of colors represented by the RGB color filter used by the image sensor. Therefore, the region c (a shadow region) including two parts which is a difference between the regions a and b is a region that can be sensed by the eye of the person but cannot be represented by the image sensor using the RGB color filter. Therefore, the conventional image sensor using the three color filters has a problem in that a region of colors that cannot be represented exists.